Chez Chesler

Crime & Federalism was persuaded by Alabama Attorney General Troy King's brief that it ought not to be unconstitutional under the 8th Amendment to apply the death penalty in any cases where the murderer was 16 or 17 years old at the time of the murder. (Thanks as always to The Volokh Conspiracy's Orin Kerr, for the pointer.) King's brief does not persuade me. I'd probably be willing to execute most of those kids myself, personally, but that shouldn't be the legal standard. (I'm wobbly on the death penalty in general, and mostly settle on the notion that it should exist, but that any court that pronounces that sentence more than once in fifty years is probably too harsh.) I know that for myself, I had a pretty good moral compass most of my life. (In fact, it's become more situationally malleable as I've aged.) But I know that at age 17 I didn't comprehend the world as well as I do now. King makes a reasonable point about bright lines, but he has failed to show that any of these murderers have the full wisdom and maturity of an adult, merely that they are capable of doing evil, capable of considerable planning in doing the evil, and capable of realizing that society will punish them if they get caught. A dog knows not to soil the carpet, and it may even try to cover it up, but it doesn't know right from wrong, it only knows that certain unpleasant consequences follow getting caught from taking certain actions. {PETULANT_TEENAGER_MODE ON} Is King suggesting that 16- and 17-year-olds, or at least those capable of plotting murders, who would face the responsibilities of adults due to their capabilities, be given the rights of adults (to vote, to contract, to drive, to imbibe, etc.) as well? {PETULANT_TEENAGER_MODE OFF}

In late September, IBM announced that its new Blue Gene/L system was the world's fastest computer, at 36.01 teraflops (3.6 times 10^13 floating-point operations per second.)

Some four years after Moore's Law was formulated, I read in My Weekly Reader about some particularly large and powerful computer. At the time, "scientists" answered not the question When will computer hardware match the human brain? but instead how large such a computer would be. If I remember correctly, it would be as large as a gymnasium, and would consume enough power to light a small city. Moravec 1998 op. cit. suggests that a 100-MIPS robot is equivalent in intelligence to a housefly, and estimates 100 million MIPS (10 ^ 14 integer operations per second) for human intelligence -- I have no idea what those "scientists" used. (Moravec notes that wetware is much more adaptable to a variety of tasks; dedicated hardware has long been faster at particular tasks.)

In any case, what was the largest computer ever built? Not in processor speed or memory, but in units of volume and electric power, gymnasia and small cities?